Skip to main content Skip to navigation

An Augmented Reality Platform for Introducing Reinforcement Learning to K-12 Students with Robots

Project Overview

The document explores the application of generative AI in education, particularly through an innovative Augmented Reality (AR) platform aimed at K-12 students. This platform facilitates the introduction of Reinforcement Learning (RL) concepts in an engaging manner by allowing students to interact with and assist in training robots, thereby visualizing the robot's learning process. By fostering an immersive learning environment, the AR system enhances students' comprehension of complex AI concepts, particularly in STEM areas. The findings suggest that such interactive experiences not only make difficult subjects more accessible but also stimulate students' interest and motivation in learning about artificial intelligence and technology. Overall, the use of generative AI through AR in educational settings demonstrates promising outcomes for improving student engagement and understanding of crucial AI principles.

Key Applications

Augmented Reality platform for Reinforcement Learning

Context: K-12 education, targeted at middle and high school students

Implementation: The AR application is installed on mobile devices, interfacing with LEGO SPIKE Prime robots to visualize the robot's learning process in a treasure hunting activity.

Outcomes: Enhanced understanding of key RL concepts such as state, action, and reward; improved engagement and interaction in learning AI; reduced complexity in training robots.

Challenges: Limitations of physical robots in visualizing AI training processes; ensuring intuitive user interaction with the AR interface.

Implementation Barriers

Technical barrier

Physical robots lack intuitive visualization methods for AI training processes.

Proposed Solutions: Utilize AR to provide visual aids that represent the robot's learning process and allow students to interact with the robot's training.

Engagement barrier

K-12 students may find it challenging to understand complex AI concepts.

Proposed Solutions: Design engaging activities around games and visualizations to simplify the learning experience.

Project Team

Ziyi Zhang

Researcher

Samuel Micah Akai-Nettey

Researcher

Adonai Addo

Researcher

Chris Rogers

Researcher

Jivko Sinapov

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Ziyi Zhang, Samuel Micah Akai-Nettey, Adonai Addo, Chris Rogers, Jivko Sinapov

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies